modelscope / modelscope/ms-swift

monitor training efficiency performance metrics in logs and tensorboards

Open
#7,012 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

stale
Dominant language
Python
Stars
15.7k
Forks
1.7k
Avg merge
1d 16h
Merged PRs (30d)
136

Description

How do you guys generally monitor the key training performance metrics of the model during training?
For example, tgs, MFU, etc.
I feel that there can be a module that can support mainstream models and custom extensions

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

The issue does not identify implementation files, entry points, or tests. Start by locating the training metric and logging integrations for mainstream models, then review how custom extensions could be supported. Done should include a defined module that records metrics such as tgs and MFU in logs and TensorBoard.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning, observability
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.